GexLab v2

Most retail-facing options dashboards either:

• stop at raw chain data,
• flatten everything into one noisy screen,
• or hide the actual structural map behind generic charts.

GexLab v2 is built around a different idea:

• show dealer positioning clearly,
• separate workflows into focused pages,
• keep replay and live mode in the same product,
• make important levels portable into TradingView,
• and preserve a high-signal studio feel instead of a spreadsheet feel.

• Track net GEX, DEX, vega, charm, and strike-level pressure.
• Surface gamma flip, call wall, put wall, session ceiling, session floor, max pain, vanna magnet, and more.
• Inspect OI walls, weak OI pockets, protected gamma ranges, aggressive flow levels, and skew-rich / skew-cheap strikes.
• Filter relevant levels by DTE for near-expiry structure up to 5 days.

• Poll live SPY / QQQ data during active sessions.
• Save EOD-style snapshots to local storage under data/snapshots/....
• Jump to historical dates and replay the same analytics stack off saved files.
• Fall back to saved data overnight instead of leaving the UI empty.

• Switch between ETF native and futures converted pricing.
• Use dark / light mode across the full studio.
• Move through dedicated pages instead of one overloaded dashboard.
• Reorder sidebar navigation by dragging and dropping items.
• Keep per-page scroll position when switching routes.

• Export a compressed TradingView payload.
• Use the bundled Pine indicator scaffold in gexlab_v2_indicator.txt.
• Carry richer relevant levels into TV instead of only a minimal wall set.

                         GEXLAB V2
        ┌──────────────────────────────────────────────┐
        │            Next.js Studio Frontend           │
        │  overview • levels • exposure • replay • TV │
        └──────────────────────┬───────────────────────┘
                               │
                     /api/metrics/*  /api/history/*
                               │
        ┌──────────────────────▼───────────────────────┐
        │              FastAPI Quant Backend           │
        │ ingestion • basis • analytics • levels       │
        └───────────────┬───────────────┬──────────────┘
                        │               │
                Yahoo / basis     Macro event sources
                        │               │
                        ▼               ▼
                local snapshots     cached event feed

data/
  snapshots/
    SPY/
      2026-04-10.json
    QQQ/
      2026-04-10.json

cd backend
venv\Scripts\python -m pip install -r requirements.txt
venv\Scripts\python -m uvicorn main:app --reload

cd frontend
npm install
npm run dev

Frontend: http://localhost:3000
Backend:  http://127.0.0.1:8000

run.bat

• GET /
• GET /api/health
• GET /api/metrics/raw
• GET /api/metrics/analytics/{ticker}
• GET /api/metrics/bridge/{ticker}

• GET /api/history/{ticker}/dates
• GET /api/history/{ticker}/{snapshot_date}

• GET /api/events/macro

backend/
  services/
    analytics/
    macro_events.py
    storage.py
  tests/
  main.py
  models.py

frontend/
  src/
    app/
    components/
    hooks/
    lib/
    types/

data/
  snapshots/
    SPY/
    QQQ/

• Next.js 16
• React 19
• TypeScript
• Tailwind CSS v4
• framer-motion
• next-themes
• Recharts
• Plotly

• FastAPI
• uvicorn
• yfinance
• pandas
• numpy
• scipy
• httpx

The UI direction for this repo is intentionally not “generic dashboard SaaS.”

It is designed to feel:

• editorial
• premium
• high-signal
• restrained in motion
• readable under heavy information density

That design guidance is also captured in .impeccable.md.

flowchart LR
    A[Raw Chain] --> B[Analytics Engine]
    B --> C[Level Intelligence]
    C --> D[Studio Pages]
    C --> E[TradingView Bridge]
    B --> F[Snapshot Storage]
    G[Macro Events] --> D
    F --> D

• focused routed experience instead of one bloated page
• local replay and user-added snapshots
• official-source macro context direction
• strong relevant-level extraction
• futures conversion workflow
• TradingView handoff path

• live options chain data currently relies on yfinance, which is practical but not a licensed direct-feed solution
• macro events are cleaner than market-news integration, by design
• local snapshot quality depends on the chain data available when saved

• add README screenshots / GIFs once the UI settles
• add a sample snapshot JSON template for user imports
• add a reset sidebar order control in-app
• add tests around more derived level families
• add a production deployment guide

Use the bundled indicator scaffold here:

• gexlab_v2_indicator.txt

Typical workflow:

1. Open the dashboard.
2. Copy the bridge payload.
3. Paste it into the indicator inputs.
4. Toggle the level families you want visible in TV.

This repo is currently structured like a private project workspace. If you plan to open-source it, the next step is to add an explicit license and a cleaned-up contribution section.